The precise nature of Roman fish sauce has until now only been imagined. It was necessarily assumed to be a clear free flowing liquid not un-similar to the modern varieties of fish sauce found in South East Asia.
These products are clear filtered liquids aptly described by Pliny as looking like ‘aged honey wine’ (HN 31.93) . How one get from piles of dissolving fish (with all that that conjurors up in our imagination) to this clean sanitized product is impossible to comprehend without empirical knowledge of the product in manufacture.
Experiments were conducted to manufacture a bulk fish sauce over the last 2 year in conjunction with the University of Reading and in part completion of an MA. In this paper I will report of the findings of these experiments as they pertain to the practicalities of trading in fish sauce products. A theory is developed as to how the nature of these products determined how they were traded. The various design features of the numerous fish sauce amphorae will be outlined and discussed in light of these findings.
Conclusions will be drawn as to the motivation behind particular design features, such as the long narrow necks of Dressel 12, more reminiscent of a wine amphora and the pronounced hollow spike in types Dressel 7-11, the logical purpose of which was clearly to hold a residue.
The residue associated with fish sauce, allec, was once considered a product of little value, to be either discarded or given to slaves (Van Neer and Ervynck 2002:208). Allec has now been observed in manufacture and it is clear that its role in the fish sauce trade is more complex and its value considerably greater than we thought.